recent developments.

• I've got a 102K (full 127031) file with exponential width correction in int5 with the standard cuts, and another with the normal width correction. These two don't differ much at all, and look much like Joao's plots for the same run in int2.
• In the histutils.h file: lots of Joao's cute macros for plotting and postscript, and my utilities for dealing with 8 superlayers.Plot eight pairs of profiles on a preformmated pad? With pleasure.
• Also, the selector now makes most of the plots we like (good tracks only, of course). Usage:

  
  root [0] TFile *f = new TFile("szt5E.root"); 
  root [1] TTree* tre = (TTree *) f->Get("SZTreeModule/SZTrackTree");
  root [2] tre->Process("cell.C+","exppost")
  Making PostScript file. Type: Exponential width correction.   
          

• Not much help from width correction plots for positive and negative tracks in new drift model (it's always possible, I guess, that the difference lies in the old drift model...)
• 50k comparisons of "my" drift model versus "joel's" are totally bizzare. Exponential gives more tracks and hits on tracks as well as smaller residuals and a totally opposite asymmetry in efficiency for positive and negative tracks. (The new pair, whole dataset, agree much better, as do the first 100k tracks in the new pair. They both look like the exponential correction of 50k.) Difference must be in a rogue package from a non-int5 release that I built into my exe.
• Preliminary version of sin_aspect vs. distance distributions for 50K events in run 127031. These come as a set of histograms, which you should right-click unless you're like that guy in the Matrix who could parse the green scrolling text.
  It's subject to
  • Funky-a width correction
  • |d0| < 1.0
  • |curv| < 0.01
  • |z0| < 60.0
  • |z120| < 100.0

<link rel="stylesheet" href="http://harv11.fnal.gov/~holloway/sz/a.css" type="text/css">